R Notebook

library(rvest)
library(tidyverse)
library(magrittr)
library(ncaahoopR)
library(sqldf)

proj = "C:/Users/Orion Wilcox/Documents/Personal/Projects/NCAA Schedules"

#Pull data

#helper data from ncaahoopsR
dict = dict
teams = ids

#rankings from NCAA website (updated daily)
ncaa = read_html("https://www.ncaa.com/rankings/basketball-men/d1/ncaa-mens-basketball-net-rankings")

net_rank = ncaa %>%
  html_element("table") %>%
  html_table() %>%
  select(Rank, School) %>%
  janitor::clean_names()

#grab vector of top 25 teams based on net ranking
top_25 = filter(net_rank, rank <=25)

top_25 = left_join(top_25, dict, by = c("school" = "NCAA")) %>%
  select(rank, school = ESPN)

top_25_list = top_25$school

#grab vector of bubble teams

bubble = filter(net_rank, rank >= 63 & rank <= 72)

bubble$school = recode(bubble$school, 
                     "Ohio St." = "Ohio State",
                     "St. Bonaventure" = "St Bonaventure")

bubble_list = bubble$school

pull schedules for top 25 teams

df = data.frame()

for (i in bubble_list) {
  sched = get_schedule(i)
  sched$team = i
  df = bind_rows(df, sched)
  
}


net_rank_dict = left_join(net_rank, dict, by = c("school" = "NCAA"))

df_rank = sqldf("SELECT * FROM df 
       LEFT JOIN net_rank_dict
       ON df.opponent = net_rank_dict.school
       OR df.opponent = net_rank_dict.ESPN
       OR df.opponent = net_rank_dict.ESPN_PBP
       OR df.opponent = net_rank_dict.Warren_Nolan")

#manually review for missing D1 schools

df$opponent = recode(df$opponent, 
                     "Fairleigh Dickinson" = "FDU",
                     "Miami OH" = "Miami (OH)",
                     "Tarleton State" = "Tarleton St.")

df_rank = sqldf("SELECT * FROM df 
       LEFT JOIN net_rank_dict
       ON df.opponent = net_rank_dict.school
       OR df.opponent = net_rank_dict.ESPN
       OR df.opponent = net_rank_dict.ESPN_PBP
       OR df.opponent = net_rank_dict.Warren_Nolan") %>%
  select(1:10) 

df_rank %<>%
  mutate(result = ifelse(team_score > opp_score, "Win", "Loss"),
         quad = case_when(location == "H" & rank <= 30 ~ "Quad 1",
                          location == "N" & rank <= 50 ~ "Quad 1",
                          location == "A" & rank <= 75 ~ "Quad 1",
                          location == "H" & rank <= 75 ~ "Quad 2",
                          location == "N" & (rank >= 50 & rank <=100) ~ "Quad 2",
                          location == "A" & (rank >= 75 & rank <=135) ~ "Quad 2",
                          location == "H" & (rank >= 75 & rank <= 160) ~ "Quad 3",
                          location == "N" & (rank >= 101 & rank <= 200) ~ "Quad 3",
                          location == "A" & (rank >= 135 & rank <= 240) ~ "Quad 3",
                          TRUE ~ "Quad 4"))

df_rank %<>%
  select(team, date, opponent, team_score, opp_score, result, quad, rank, location, team_score, opp_score) %>%
  group_by(team) %>%
  mutate(game_order = dense_rank(date))

#chart

each team’s worst loss

#export

ggsave(paste0(proj, "/visuals/top_25.png"), width = 1200, height = 1500, unit = "px")

#datawrapper export

write_csv(top_25_sched, paste0(proj, "/data/top_25.csv"))